Prevention Program for Problem Behaviors in Girls in Foster Care

36 items from the general delinquency scale from the Self-Report Delinquency Scale (SRD; Elliott, Huizinga, & Ageton, 1985). Units on a scale. Girls were asked to rate how many times they had committed various delinquent acts (e.g., damaging or destroying properties, and stealing) in the past year, using an open-ended format. The mean of frequencies across these items was used to represent the level of delinquency for girls. The general delinquency scale scores ranged from 0 to 24 (full scale) and from 0 to 13 (log transformed). Higher scores indicate higher levels of delinquency.

The girls were asked how many times in the past year they had smoked cigarettes or chewed tobacco. The response scale ranged from 1 (never) through 9 (daily). Units on a scale.

The girls were asked how many times in the past year they had used marijuana. The response scale ranged from 1 (never) through 9 (daily). Units on a scale. Log transformed.

Internalizing and externalizing symptoms at 12 and 24 months were measured with caregiver report on the Achenbach System of Empirically Based Assessment (ASEBA). This widely used checklist for psychopathological behaviors includes scales for behaviors such as Anxious/Depressed; Withdrawn; Somatic Complaints; Thought Problems; Attention Problems; Aggressive Behavior; Rule-Breaking Behavior; and Intrusive. The ASEBA has been shown to have both construct and content validity in the literature. For the present study, raw scores for the internalizing and externalizing symptoms subscales were used. Scores at 12 and 24 months were combined and averaged (mean). Units on a scale. Range = 0-66. Higher scores indicate higher levels of internalizing or externalizing problems.

Eight items from the girls' in-person interviews were used to assess health risking sexual behavior at the 36-month followup. The girls reported on items such as touching a boy's body above or below the waist, having sexual intercourse, having sex with someone who they just met, or having sex with someone using drugs in the past 12 months. Positive answers to these items were totaled to represent the cumulative number of health-risking sexual behaviors. The frequency of the cumulative number of risky sexual acts ranged from 0 to 7. Units on a scale. Higher scores indicate more health-risking sexual behaviors.

Prosocial behavior was measured with a subscale from the Parent Daily Report (PDR; Chamberlain & Reid, 1987). The PDR was administered individually by telephone to foster parents on 3 consecutive or closely spaced days (1-3 days apart) at each assessment. A trained interviewer asked the foster parent whether a list of prosocial behaviors took place during the previous 24 hr (yes/no format). The prosocial scale was computed based on nine items, such as "cleans up after herself" and "do a favor for someone." The PDR was designed to avoid the potential bias of aggregate recall of frequency estimates. Studies have reported concurrent and predictive validity of the PDR checklist. The scores were averaged (mean) across calls from 3 days. Scores on prosocial behavior at 6 and 12 months were averaged and the mean across both time points was used in analysis. Units on a scale. Range = 0-9. Higher scores indicate more prosocial behavior.

Child welfare system records were collected at each assessment to determine the girls' placement changes (including the number and type of changes). Placement changes since the start of the study through 12 months were summed for each girl. The number of placement changes ranged from 0 to 7 during this period. Units on a scale. Higher scores indicate more placement changes.

"Cups" task (Weller et al., 2007). On each trial, participants see 2 arrays with equal number of X cups (2, 3, or 5) each. On gain trials, participants informed that under each cup in one array is 1 quarter, and the other array includes 1 cup with Y quarters (either 2, 3, or 5), but the other cups have 0 quarters. Choosing from the riskless side leads to a sure gain of 1 quarter while choosing the risky side can lead to gain of Y quarters or no quarters. On loss trials, participants shown that choosing cup from 1 array will lead to 1 quarter taken away while choosing cup from other array will lead to no quarters or Y quarters taken. Cups task consists of 54 trials of 3 trials each of all combinations of 2 levels of domain (gain, loss). Expected Value Sensitivity (EV) calculated by subtracting proportion of risky choices made when EV actually favored the sure choice from proportion of risky choices made on trials where EV favored risky option. Score can range from -1.0 to -1.0.